This competing continuation requests 5 years of funding to continue our efforts to reduce the transmission of HIV among non- injection drug users at high risk for transition to injection. This application takes the innovative approach of extending intervention research to include the primary prevention of injection among heroin sniffers who have never injected drugs. A prospective experimental design with a nested ethnography is proposed. A tri-ethnic sample of 750 men and women who sniff heroin will be enrolled and randomized to two experimental interventions and one comparison group and followed at 6, 12, and 18 months to evaluate the effectiveness and sustainability of two interventions to reduce high risk sex behaviors, prevent the transition to injection and increase enrollment in drug treatment. We propose an efficient design which permits us to simultaneously implement and evaluate the effectiveness of the interventions and prospectively study the transition to injection. There are three specific aims: 1) Implement, evaluate and compare the effectiveness and sustainability of two brief HIV risk interventions among heroin sniffers. A controlled randomized experimental design will compare two distinct interventions, a Risk Prevention Intervention and a Stage-Enhanced Motivational Interviewing Intervention with a"usual pre-and post-test counseling" group. 2) Characterize and model risk factors for the transition from heroin sniffing to injection. A prospective epidemiologic design with a nested ethnography will estimate incidence of injection, model risk factors for initiation of injection, and characterize the process of transition to injection/early injection experiences of heroin sniffers who shift to injection. 3) Estimate the prevalence and incidence of HIV-1 among heroin sniffers. Standard epidemiologic techniques will be employed. Gender specific and ethnic specific estimates will be conducted.